Initialize a susiF object using regression coefficients

Initialize a susiF object using regression coefficients

Usage

init_susiF_obj(
  L_max,
  G_prior,
  Y,
  X,
  L_start,
  greedy,
  backfit,
  tol_null_prior = 0.001,
  cov_lev = 0.95,
  ...
)

Arguments

L_max

upper bound on the number of non zero coefficients An L-vector containing the indices of the

G_prior

prior object defined by init_prior function

Y

Matrix of outcomes

X

matrix of covariatess

L_start

number of effect to start with

greedy

logical, if TRUE allow greedy search

backfit

logical, if TRUE allow backfitting

tol_null_prior

threshold to consider prior to be null. If the estimated weight on the point mass at zero is larger than 1-tol_null_prior then set prior weight on point mass to be 1. In the mixture normal this corresponds to removing the effect. In the mixutre per scale prior this corresponds to setting the prior of a given scale to at point mass at 0.

cov_lev

numeric between 0 and 1, corresponding to the expected level of coverage of the CS if not specified, set to 0.95

...

Other arguments.

Details

Initialize a susiF object using regression coefficients